15th International Science of Aphasia Conference, September 2014

Stem- Spraak- en

Taalpathologie

Supplement, September 2014

15

th

_{International Science of Aphasia}

i

PREFACE

Dear participants,

We are very pleased to welcome you to the 15th _{Science of Aphasia conference, being}

held from September 19 till September 24, 2015 in the San Camillo Hospital in Venice, Italy

The 2014 program theme is:

Aphasiology: past, present and future

Invited speakers are: Ria De Bleser, Audrey Bowen, Marco Catani, Chris Code, Olga Dragoy, Hugues Duffau, David Howard, Peter Mariën, Gabriele Miceli, Carlo Miniussi, Lyndsey Nickels, Carlo Semenza, Cynthia K. Thompson, Evy Visch-Brink, Frank Zanow.

The SoA conferences are intended to bring together senior and junior scientists working in the multidisciplinary field Neurocognition of language and to deal with normal

function as well as disorders. The size of the conference has a maximum of about 150 participants to ensure direct interaction between the participants. The focus of this year’s conference is on the past, present and future of Aphasiology:

The San Camillo Hospital in Venice-Lido is a health care facility, mainly devoted to the rehabilitation outcomes of traumatic brain injury and spinal cord, stroke, multiple sclerosis, amyotrophic lateral sclerosis, Parkinson's disease, neuropathy and dementia. In 2005 the hospital received recognition from the Ministry of Health of the Institute for Research, Hospitalization and Health Care (IRCCS) specializes in the "discipline of neuro-rehabilitation motor, communication and behavior." The experience in telemedicine, robotics and Brain Computer Interface (BCI) allowed the hospital to develop a communication system based exclusively on the modulation of brain activity recorded with an electroencephalograph, even without moving a muscle. This system will allow people not able to perform movements or to speak to communicate and also to carry out activities. The San Camillo Hospital is situated on the Lido of Venice. The Lido — or Venice Lido (Lido di Venezia) — is an 11 kilometres (7 miles) long sandbar in Venice; it is home to about 20,000 residents.

The city of Venice, or in Italian Venezia, is a city in northeastern Italy, sited on a group of 118 small islands separated by canals and linked by bridges. It is located in the marshy Venetian Lagoon which stretches along the shoreline, between the mouths of the Po and the Piave Rivers. Venice is renowned for the beauty of its setting, its architecture and its artworks. The city in its entirety is listed as a World Heritage Site, along with its lagoon. We wish you a pleasant conference!

ii

Organization

The 15th International Science of Aphasia Conference is held in Venice, Italy, September 19-24, 2014.

Chair:

Prof. Carlo Semenza, University of Padova, Italy

The 2014 scientific committee is composed of: Ria de Bleser (honorary member)

Roelien Bastiaanse (chair) Wendy Best Frank Burchert David Howard Roel Jonkers Gabriele Miceli Lyndsey Nickels Brendan Weekes

The 2014 abstract selection committee is composed of: Roel Jonkers (chair)

Frank Burchert David Copland David Howard Lyndsey Nickels Isabell Wartenburger Abstract Booklet Alice Pomstra

iii

Conference Program

Friday, September 19, 2014

17:30 – 19:00 Reception and registration

Saturday, September 20, 2014

(chair Roelien Bastiaanse)

9:30 Roelien Bastiaanse: Intro: Pre Broca

9:45 Chris Code : Broca and his contemporaries

10:30 Ria de Bleser: German aphasiology

11:15 Coffee Break

11:45 Marco Catani: American aphasiology (Boston School)

12:30 Olga Dragoy: Russian aphasiology (Luria)

13:15 – 15:00 Lunch

15:00 – 17:00 Contributed Papers I (Chair Olga Dragoy)

Bos.: The neural correlates of past time reference

Arslan: Source memory deficits in aphasic and healthy aging speakers of

Turkish

Popov: Unaccusative Verb Production Revisited: Evidence for Dual Deficit Salmons: The comprehension of Catalan OVS and OSV structures in Broca’s

aphasia

Jochmann: The effects of slowed speech on comprehension of German

non-canonical sentences in aphasics with and without hearing impairment

Munarriz: The role of typological distance in differential impairments in

bilingual aphasia: evidence from Spanish-Basque agrammatism. 17:00 – 17:30 Short presentations Poster Session I

(Chair Sylvia Martinez-Ferreiro) 17.30 – 18.30 Poster Session I/Coffee

Adelt: Do Pronouns Make a Difference? On-line Processing of Relative

Brandao: Communicative strategies in expressive aphasia: discourse as a

guideline for rehabilitation

Capitani: Lexical-semantic errors are more consistent than phonological

errors on the repeated naming of the same picture: a study on aphasic patients.

Feiden: Anomia and paraphasia in oral speech production

Fyndanis: Structural case in agrammatic aphasia: Evidence from Greek Gora.: Cross-language influences in multilingual aphasia

Haaland: “I wake up every day thinking I can write” – an agraphia

treatment study

Ishkanyan: Syntactic comprehension deficits in Armenian-Russian

bilingual speakers with aphasia

Jap: Sentence Comprehension in Aphasic Speakers of Standard Indonesian Khakalo: A Vaster VAST: Comprehension and production of verbs and

sentences in Russian

Knoph: The impact of Semantic Feature Analysis on verb production in two

multilingual speakers with aphasia

Satoer: Glioma surgery in eloquent areas, can we preserve cognition?

Sunday, September 21, 2014

Session Two. Presentations I: Awake surgery (Chair: Evy Visch-Brink)

9:30 Hugues Duffau: Perspective from a neurosurgeon

10:15 Peter Mariën: Perspective from a neurolinguist

11:00 – 11:30 Coffee Break

11:30 Carlo Semenza: Perspective from a numberologist

12:15 General discussion 13:00 – 15:00 Lunch

15:00 – 17:00 Contributed papers II (Chair Davide Crepaldi)

De Witte: Non-Organic Language Disorders after Awake Brain Surgery Groenewold:: The effects of direct and indirect speech on English discourse

comprehension in aphasia

Romanova, et al: Facilitation effect in proper and common noun naming Bose: Relationship between semantic transparency of compound words

and semantic processing skills in aphasia: Data from compound word reading

Ribu: Imageability and phonological neighborhood density effects in

speech processing

Keulen: Foreign Accent Syndrome: a typological overview

17:00 – 17:30 Short presentations Poster session II (Chair Sylvia Martinez-Ferreiro)

Bambini: The Italian version of the Communication Outcome after Stroke

(COAST) scales for patients and caregivers

Jesus: Assessment of Aphasia in Portugal: Past, present and future

Manouilidou: Lexical-semantic deficits in Mild Cognitive Impairment, et al:

the case of abstract vs. concrete nouns

Pellet: Outcome of computer-assisted treatment in a case of non-fluent

primary progressive aphasia with apraxia of speech

Penaloza: Associative learning and retention of novel labels for novel

visual referents in patients with chronic aphasia

Stavrakaki: Production of verbs with alternating transitivity by patients

with Primary Progressive Aphasia

Vlasova: Speech disorders and its postoperative progress in patients with symptomatic epilepsy

Wimmer: Verbal Agreement Inflection in Wernicke’s and Broca’s Aphasia –

a comparison

Zanini: When verbs help naming nouns: a study on derived nominals in

aphasia

Azimova: Verbs in Uzbek agrammatic spontaneous speech

Zivanovic: Predictors of Post-Stroke Aphasia Recovery – A Systematic

Review

Vukovic: Quality of communication life in individuals with Broca’s and

Conduction aphasia

Lesniak: Language dissolution and restitution in L1 and L2

18:30 Meeting Scientific Committee

Monday, September 22, 2014

Presentations II: Aphasia Trials

(

9.30 Audrey Bowen: Randomized Control Trials: The ActNow study 10.15 Evy Visch-Brink: Randomized Control Trials: The RATS studies 11.00 Coffee break

11.30 Lyndsey Nickels: Single Subject Experimental Design 12.15 David Howard: Single Subject Experimental Design and

Randomized Control Trial: The Semafor study 13.00 Lunch, Excursion and Dinner

vi

Tuesday, September 23, 2014

(Chair: David Howard)

9.30 Frank Zanow: Combination of neuro imaging methodologies 10.15 Carlo Miniussi: Transcranial Direct Current Stimulation 11.00 Coffee

11.30 Gabriele Miceli: The use of transcranial Direct Current Stimulation in aphasia therapy

12.15 Cynthia K. Thompson: Neuro Imaging in the study (and recovery) of aphasia

13.00 - 15.00 Lunch

15.00 – 17.00 Contributed papers III (Chair Roel Jonkers)

Crepaldi: How to become twice more precise in detecting

neuropsychological impairments

De Aguiar: Aphasia rehabilitation from a linguistic perspective and the role

of tDCS

Hanne: An eye tracking study of time reference processing in individuals

with agrammatic aphasia

Silberling: Eye movement based evaluation of a text-level reading

intervention

Krajenbrink: Investigating the role of neighbours in treatment of acquired

dysgraphia

Rofes: Naming finite verbs predicts language abilities in daily living

17.00 – 17.30 Short presentations Poster session III (Chair Roel Jonkers)

17.30 – 18.30 Poster session III/ coffee

Yarbay Duman: Tense, Aspect and Modality in Three Populations:

Typically Developing Children, Children with Specific Language Impairment (SLI) and individuals with Broca's Aphasia

Martinez Ferreiro/Reyes: Strategies in non-fluent aphasia: Object clitic

substitutions and redundant structures

Pistono: Memory and language interactions during discourse of patients

with MCI due to AD

Lange: Speech Mapping of the Broca Region using repetitive Transcranial

vii

Primassin: Neural Correlates of Motor and Language Recovery after

Stroke: Four Single Case Studies

Martin: Improving communication in aphasia: A comparison of naming-

and discourse based treatment, both facilitated by transcranial direct current stimulation (tDCS)

Van Dun: Cerebellum and Apraxia

Zakarias: Positive effects of computerized executive function training in

aphasia. A pilot study

Meteyard: Sentence production training in severe expressive aphasia and

apraxia of speech: a case study

Roxbury: Relationship between subacute brain activity and aphasia

recovery

Bambini: Concretism, pragmatics, and the interplay of language and

cognition in schizophrenia

Franzon: Semantic interpretability speeds up the processing of

morphologic features. A psycholinguistic experiment on Gender Agreement

Pourquie: Past, Present and Future of Basque aphasiology and

Cross-linguistic studies of aphasia

Sierpowska: Mapping the dorsal and ventral language streams using

electrical stimulation and diffusion tensor imaging.

Wednesday, September 24, 2014

The neural correlates of past time reference

Laura S. Bos1,2,3_{, Roelien Bastiaanse}2_{, Jan Ries}3_{, and Isabell Wartenburger}3

1_{International Doctorate in Experimental Approaches to Language and Brain (IDEALAB)} 2_{University of Groningen (Groningen, NL)}

3_{University of Potsdam (Potsdam, GER)}

Introduction

Background

Tense inflection is problematic for individuals with agrammatic aphasia. Recently, it has been claimed that past time reference by verb inflection is particularly impaired,

because it requires additional processing: In order to use a verb with past time reference, one has to make a link to the event in discourse. No such link is needed for non-past time reference. Therefore, non-past verb forms are relatively spared in agrammatism. This is captured in the Past DIscourse LInking Hypothesis (PADILIH; Bastiaanse et al., 2011).

The PADILIH has been argued to extend to all discourse-linked verb forms, irrespective of tense. Crucially, tense does not always coincide with time reference. In Dutch,

reference to the past can be established by a finite verb in the simple past tense (past imperfect), or by a periphrastic past verb form (the present perfect). The latter form consists of an auxiliary in present tense plus a participle (i.e. ‘has+V-ed’). Dutch aphasic speakers are impaired on both types of past time reference compared to the simple present (Bos & Bastiaanse, 2014), with the simple past being the most difficult form to produce.

The PADILIH is a processing-based hypothesis of which traces can be found in non-brain-damaged speakers, too. In their electrophysiological brain responses, discourse-related differences between past and non-past time reference have been found (Dragoy et al., 2012), which cannot be related to tense per se (Bos et al., 2013).

Current study

The PADILIH suggests that reference to the past requires discourse linking and discourse linking requires extra processing; the question is which brain regions are engaged in this additional processing. Multiple fMRI studies have shown that the left inferior frontal gyrus (IFG) is essential for processing of grammatical morphology (e.g., Sahin et al. 2009) and that in agrammatic aphasia the left IFG is often not functional (e.g., Damasio, 1991). Hence, IFG is a likely candidate for the extra processing needed for past time reference. Another possibility is that the right hemisphere plays a role in discourse linking needed for reference to the past. Malfunctioning of the left IFG may cause a disconnection with the contralateral area, which has been associated with discourse processing by Menenti et al. (2009). Lastly, the supplementary motor area (SMA) has been associated with greater processing complexity in selecting grammatical inflections (e.g., Sahin et al., 2009; Yu et al., 2013) and is thus a candidate for additional activation in the more difficult past conditions.

In the current study we investigated whether additional brain activation is found for past versus non-past time reference for the contrast of (1) simple past versus present

tense (refuelled-refuels) and of (2) periphrastic past versus future time reference (has refuelled – will refuel). We focussed our discussion of the results on the IFG and the SMA.

Methods

Participants and Materials

Twenty healthy speakers of Dutch (7 male, mean age 24, range 19-32) participated in this fMRI-study. Thirty sentences per experimental condition (see examples 1-4 below; target responses in square brackets) were used. Participants were shown a picture denoting the action of the verb for 2 seconds. Then, the sentence was presented in written form until the penultimate phrase. One or two ellipses indicated whether they had to overtly respond with a single or periphrastic verb form, respectively. The next trial began after a jittered inter-stimulus-interval.

(1) simple present:

ik verwar de vrouw terwijl ze diesel … [tankt]

I confuse the woman while she gas … [refuels] (2) simple past:

ik verwarde de vrouw terwijl ze diesel … [tankte]

‘I confused the woman while she gas … [refuelled]’ (3) periphrastic future:

ik verwar de vrouw voordat ze diesel … … [gaat tanken]

‘I confuse the woman before she gas … … [will refuel]’) (4) periphrastic past:

ik verwar de vrouw nadat ze diesel … … [heeft getankt]

‘I confuse the woman after she gas … … [has refuelled]’)

Data acquisition and analysis

The Siemens 3T TrioTim scanner of the Dahlem Institute for Neuroimaging of Emotion (D.I.N.E.) was used to acquire functional/anatomical scans. DARTEL-normalised

functional scans of response preparation from picture offset to response onset were analysed with SPM8 (Wellcome Department of Cognitive Neurology, London, UK). Main areas of activation with a cluster size threshold of 10 are reported with uncorrected p < .001.

Results and discussion

Past over non-past

The preliminary fMRI analysis (see Figure 1) showed showed a greater BOLD signal change for both types of past over non-past time reference bilaterally in the prefrontal area – the supplementary motor area and the frontal superior medial gyrus. These prefrontal areas have been associated with increased difficulty in implementing

grammatical inflections (e.g., Sahin et al., 2009; Yu et al., 2013). This can be explained by

the PADILIH: the past is more marked than the non-past, resulting in increased difficulty in selecting past time reference inflection.

Simple past over present

The contrast of simple past (most impaired in Dutch agrammatism) and present

additionally yielded increased activation in the left inferior frontal gyrus, the area which is commonly compromised in agrammatism. The lack of significant activation on the left IFG for the contrast of periphrastic past over future suggests that this region is taxed for past tense inflection, but not for discourse-linked past time reference specifically. Even though the periphrastic past and the future have different time reference, they are both verb clusters with a present tense auxiliary.

Overall, the results provide further support for a fundamental difference between past and non-past time reference, and reflect processing difficulties that aphasic individuals experience with past time reference.

-FIGURE-1-ABOUT-HERE-

Figure 1. Activation patterns for (A) simple past over simple present and (B)

periphrastic past over future (uncorrected, p < .001). Legend: IFG = inferior frontal gyrus; SMA = supplementary motor area.

References

Bastiaanse, R., Bamyaci, E., Hsu, C., Lee, J., Thompson, C.K., & Yarbay-Duman, T. (2011). Time-reference in agrammatic aphasia: A cross-linguistic study. Journal of

Neurolinguistics, 24, 652-673.

Bos, L.S., & Bastiaanse, R. (2014). Time reference decoupled from tense in agrammatic and fluent aphasia. Aphasiology, 28, 533-553.

Bos, L.S., Dragoy, O., Stowe, L.A. & Bastiaanse, R. (2013). Time reference teased apart from tense: Thinking beyond the present. Journal of Neurolinguistics, 2, 283–297. Damasio, H. (1991). Neuroanatomical correlates of the aphasias. In: M. Sarno, ed.

Acquired aphasia. 2. San Diego: Academic Press; pp. 45–71.

Dragoy, O., Stowe, L.A., Bos, L.S., & Bastiaanse, R. (2012). From time to time: Processing time-reference violations in Dutch. Journal of Memory and Language, 66, 307-325. Menenti, L., Petersson, K. M., Scheeringa, R., & Hagoort, P. (2009). When elephants fly: Differential sensitivity of right and left inferior frontal gyri to discourse and world knowledge. Journal of Cognitive Neuroscience, 21, 2358-2368.

Sahin, N.T., Pinker, S., Cash, S.S., Schomer, D., & Halgren, E. (2009). Sequential Processing of Lexical, Grammatical, and Phonological Information Within Broca’s Area. Science, 326, 445-449.

Yu, X., Bi, Y., Han, Z., & Law, S.-P. (2013). An fMRI Study of Grammatical Morpheme Processing Associated with Nouns and Verbs in Chinese. PLoS ONE 8, e74952. doi:10.1371/journal.pone.0074952.

Source memory deficits in aphasic and healthy aging

speakers of Turkish

Seçkin Arslan1 _{and Roelien Bastiaanse}2

1 _{International Doctorate for Experimental Approaches to Language and Brain (IDEALAB),}

University of Groningen, The Netherlands; University of Potsdam, Germany; University of Newcastle, The UK; University of Trento, Italy; and Macquarie University, Sydney,

Australia.

2 _{Center for Language and Cognition Groningen (CLCG), Department of Linguistics,}

University of Groningen, The Netherlands

Introduction

Source memory or source monitoring refers to the processes of encoding, storing, and retrieving from where and which sources a piece of particular memory was gained (Johnson, Hashtroudi, & Lindsay, 1993). Neuropsychological studies on frontal lobe patients and healthy aging adults have shown that the source memory is vulnerable (Glisky, Polster, & Routhieaux, 1995; Janowsky, Shimamura, & Squire, 1989; Swick, Senkfor, & Van Petten, 2006). However, few studies tested patients or aging adults in languages other than English. Turkish grammar, for instance, requires speakers to indicate whether past events were ‘seen’ directly or ‘heard’ through a third-person report. A recent study has shown that Turkish agrammatic speakers were more impaired in producing ‘seen past’ than ‘heard past’ verb forms (Arslan, Aksu-Koç, Maviş, & Bastiaanse, subm.). However, identifying whether the information was received through the ‘seen’ sources was better than through the ‘heard’ sources, suggesting that the agrammatic speakers might not monitor the information sources mapped onto seen/heard past verb forms accurately. Therefore, the current study addresses whether monitoring the sources of memories is impaired in the aphasic patients and healthy aging speakers of Turkish.

Methods

Participants

We tested three groups of participants: eight aphasic patients (three fluent patients: two females, Mage = 55.6; five non-fluent patients: one female, Mage = 59; post-onset > eight months), fifteen younger non-brain-damaged individuals (NBDs; nine females, Mage = 26.4), and five elderly healthy aging NBDs (4 females, Mage = 54.2).

Materials and Procedure

The source monitoring task in Arslan, de Kok, and Bastiaanse (in progress) was used. Stimuli included eighty meaningful inanimate objects. Forty objects were represented as black-white line drawings (seen items). The other forty objects were audio recorded as spoken words (heard items). All the stimuli were concrete nouns. The experimental

design included two phases: ‘study’ and ‘test’. At the study phase, all eighty objects were presented. The seen and heard items were presented as separate blocks counter-balanced across the participants. At the test phase, a subset of forty items from the study list (twenty seen items and twenty heard items) was presented as ‘old items’. An additional set of forty items was used as ‘new items’. All the items in the test phase were presented as written words. For each test item, the participants had to respond to ‘old/new’ and ‘source monitoring’ judgments. First, they had to judge whether a test item was from the study list or not. This is referred to as ‘old/new recognition’. If the participants judged an item ‘old’, subsequently, they had to decide whether it was a seen or heard item. This is referred to as ‘source monitoring’. Source monitoring responses were scored when the items were recognized correctly in the old/new judgment.

Results

Old/new recognition

Non-parametric tests were used to analyze the data. The aphasic individuals scored lower in old/new recognition than the elderly NBDs (z = -3.437, p = .001), and the younger NBDs (z = -6.221, p > .001), see Table 1. The elderly NBDs did not differ from the younger NBDs (z = -1.566, p = .117). Rejection of new items was more difficult than recognition of old items for the non-fluent aphasic individuals (X2 _{= 15.068, df = 1, p >} .001); and for the elderly NBDs (X2 _{= 6.273, df = 1, p = .012). In the fluent aphasic} individuals, however, there was not such a difference (X2 _{= .311, df = 1, p = .577). The} younger NBDs showed the opposite pattern of the non-fluent aphasic individuals and elderly NBDs: they were better in rejecting the new items than recognizing the old items (X2 _{= 24.624, df = 1, p > .001).}

Source monitoring

Source monitoring performance of the aphasic individuals was reduced compared to the elderly NBDs (z = -3.418, p = .001) and the younger NBDs (z = -7.347, p > .001). Comparably, the elderly NBDs performed lower than the younger NBDs (z = -2.547, p = .011), see Table 1. Recalling source was better attained for the seen than for the heard items in non-fluent aphasia (X2 _{= 57.063, df = 1, p > .0001); in fluent aphasia (X}2 _{= 8.120,}

df = 1, p = .004); and in the elderly NBDs (X2 _{= 6.033, df = 1, p = .014). The younger NBDs,} however, did not differ in recalling sources for seen or heard items (X2 _{=1.286, df = 1, p =} .257).

Discussion

We showed that old/new and source monitoring judgments were affected in fluent and non-fluent aphasia compared to the younger and elderly NBDs. The elderly NBDs also differed from the younger controls in making source monitoring but not old/new judgments. The non-fluent aphasic individuals were more affected in making source monitoring judgments for the ‘heard’ items than for the ‘seen’ items, although the results of the fluent aphasic individuals and elderly NBDs pointed to the same direction. Our

results, thus, confirm earlier neuropsychological studies, and indicate that lesions leading to fluent and non-fluent aphasias, as well as healthy aging, affect source monitoring judgments. Our data indicate a similar direction of source identification deficits as in Arslan et al. (subm.) who showed the agrammatic speakers of Turkish cannot attribute information sources to the seen/heard past verb forms. We argue that source confusions in Turkish non-fluent aphasia may be due to source memory deficits.

References

Arslan, S., Aksu-Koç, A., Maviş, I., & Bastiaanse, R. (subm.). Finite verb inflections for evidential categories and source identification in Turkish agrammatic Broca’s aphasia.

Arslan, S., de Kok, D., & Bastiaanse, R. (in progress). Processing evidentiality and time reference in monolingual and bilingual speakers: An auditory sentence processing and source monitoring study in Turkish.

Glisky, E. L., Polster, M. R., & Routhieaux, B. C. (1995). Double dissociation between item and source memory. Neuropsychology, 9(2), 229.

Janowsky, J. S., Shimamura, A. P., & Squire, L. R. (1989). Source memory impairment in patients with frontal lobe lesions. Neuropsychologia, 27(8), 1043-1056.

Johnson, M. K., Hashtroudi, S., & Lindsay, D. S. (1993). Source monitoring. Psychological Bulletin,

114(1), 3-28.

Swick, D., Senkfor, A. J., & Van Petten, C. (2006). Source memory retrieval is affected by aging and prefrontal lesions: Behavioral and ERP evidence. Brain Research, 1107(1), 161.

Table 1. Mean recognition and source monitoring accuracies (SDs) of the participants Old/new recognition Source monitoring Old New Total Seen Heard Total Younger NBDs .71(.45) .83(.37) .77(.41) .89(.31) .86(.35) .88(.33) Elderly NBDs .79(.40) .68(.46) .74(.43) .86(.34) .70(.46) .79(40) Non-fluent aphasia .70(.45) .51(.50) .61(.48) .74(.37) .19(.39) .54(.50) Fluent aphasia .68(.47) .71(.45) .69(.46) .91(.29) .65(.48) .79(.41) Aphasia total .69(.46) .58(.49) .64(.48) .86(.35) .36(.48) .63(.48) 7

Unaccusative Verb Production Revisited: Evidence for Dual Deficit

Srđan Popov1,2,3_{, Mile Vuković}4_{, Roelien Bastiaanse}3

1 _{International Doctorate in Experimental Approaches to Language And the Brain (IDEALAB),} 2 _Center

for Mind/Brain Sciences (CIMeC) and Center for Neurocognitive Rehabilitation (CeRIN), University of Trento, Rovereto, Italy, 3 _{Center for Language and Cognition Groningen (CLCG), University of}

Groningen, Groningen, The Netherlands,4 _{University of Belgrade - Faculty of Special Education and}

Rehabilitation, Belgrade, Serbia

Introduction

Previous studies have identified a number of factors that influence verb production in aphasia. According to the Argument Structure Complexity Hypothesis (ASCH; Thompson 2003), an increase in syntactic complexity adds to the processing load, which in turns impedes verb production. The number of arguments is recognized as the major component of syntactic complexity, predicting that verbs with more arguments are more difficult to produce. Regarding the locus of the deficit in the speech production model (Levelt 1989), the ASCH predicts that the lemma access level is impaired in verb production. Since lemmas contain the information on the number of arguments, the more arguments there are, the more syntactically complex the lemma becomes.

A study by Bastiaanse and Van Zonneveld (2005) found that the number of arguments may not be the most important grammatical factor in verb production. When comparing the production of verbs with alternating transitivity, Bastiaanse and Van Zonneveld demonstrated that the intransitive realization of the verb (1) was more difficult than the transitive one (2).

(1) John broke the vase. (2) The vase broke.

The intransitive realization in (1) is unaccusative, meaning that the only argument is the Theme that has moved to the subject position. Based on the movement property of unaccusative verbs,

Bastiaanse and Van Zonneveld concluded that the source of difficulty was in violating the basic word order, and not the number of arguments. Their assumption is that every language has a canonical underlying word order (e.g., V-S for sentences with unaccusative verbs in Dutch), and that any changes in that word order will affect production in aphasia, thus the Derived Order Problem Hypothesis (DOP-H). Unlike the ASCH, the DOP-H places the deficit in the grammatical encoding phase.

The current study looks deeper into the issue of unaccusative production and deficit localization. The previous studies have not taken into account the difference between non-derived unaccusative verbs (e.g., fall, arrive), and unaccusative verbs derived from transitive verbs (henceforth anticausatives). Theoretically, the two unaccusative classes differ at the presyntactic (lemma) level. Reinhart (2000) in her Theta System theory suggests that verbs with alternating transitivity represent one concept, and are therefore a single lexicon entry. Just before the derivation (grammatical encoding), the transitive lemma of a verb with alternating transitivity is rendered intransitive unaccusative through the lexical operation Expletivization. After that, at the grammatical encoding level, unaccusative and

anticausative verbs cannot be differentiated any more, as they both comprise a Theme movement to the subject position. This study hypothesizes that Expletivization is a factor adding to syntactic complexity of the anticausative verb. If the production of anticausative verbs is more impaired, the deficit distinguishing the two unaccusative classes should be at the lemma level, which is where the anticausative derivation takes place. Such an assumption is in line with the ASCH and syntactic complexity. If there is no discrepancy in the production between the two unaccusative classes, the DOP-H is correct in predicting that the only relevant deficit is at the grammatical encoding level.

Methods

The aphasic group consisted of four people with Broca’s aphasia, diagnosed with the Serbian adaptation of the Boston Diagnostic Aphasia Examination (Goodglass and Caplan 1972). The control group consisted of eight participants, matched for age and education level with the experimental group. All participants were monolingual speakers of Serbian.

The participants performed a sentence production task. For this purpose, 50 verbs were chosen. In order to elicit the sentences, 50 black-and-white drawings were used. Each drawing depicted one event. In addition to the visual presentation of an event, the infinitive form of the verb was given above the drawing to avoid interference with word-finding difficulties. The test consisted of 50 test items, with 10 items in the unaccusative group, 10 items in the anticausative group, and 30 items divided between one intransitive and two transitive groups that served as control conditions.

Results

The control group performed at ceiling, and was significantly better than the experimental group in all the conditions (chi-squared test: χ2_{= 79.6, df= 4, p= 0.000). The question in this study was if the} production of unaccusative verbs differed significantly from the production of anticausative verbs. The data obtained from the four people with aphasia point in this direction; anticausative verbs were significantly more impaired than unaccusative verbs (χ2_{= 5.409, df= 1, p= 0.02). The other pairwise} comparisons between the conditions are given in Table 1.

Table 1: P-values (chi-square test) for the pairwise comparisons between the conditions in the experimental group Comparison Results (p) Intransitive - Unaccusative 0.018* Intransitive - Anticausative 0.000* Unaccusative – Anticausative 0.02* Unaccusative – Transitive 1 0.22 Unaccusative – Transitive 2 1 Anticausative – Transitive 1 0.000* Anticausative – Transitive 2 0.01* Transitive 1 – Transitive 2 0.35

Discussion

The results clearly show that the production of anticausative verbs is more impaired than the production of unaccusative verbs. The results are in line with the ASCH (Thompson 2003) and show that syntactic complexity is a factor in verb production. The additional lexical operation applied at the lemma level of anticausative verbs adds to their syntactic complexity, and makes them more difficult than unaccusative verbs. The deficit, in this case, has to be at the lemma level.

Still, the results do not necessarily contradict the DOP-H (Bastiaanse and Van Zonneveld 2005). The movement, and consequently the variation in the basic word order, clearly play a role in aphasic verb production. Our data show that unaccusative verbs are more difficult than unergative verbs. Since they have an equal number of arguments, the only difference is at the grammatical encoding level where the movement operation takes place. Therefore, unaccusatives, as well as anticausatives, indicate a deficit at the grammatical encoding level. The difference between the two unaccusative classes is that anticausative verbs have an additional deficit at the lemma level, which makes them even more difficult. The difference at the lemma level can only be captured by the ASCH and syntactic complexity since the DOP-H does not have predictions about the lemma level.

References

Bastiaanse, R. and Van Zonneveld, R. (2005). Sentence production with verbs with alternating transitivity in agrammatic Broca’s aphasia. Journal of Neurolinguistics,18, 57–66.

Goodglass. H. and Kaplan, E (1972). The Assessment of Aphasia and Related Disorders. Philadelphia, PA: Lea and Febiger.

Levelt, W. (1989). Speaking. Cambridge: MIT Press.

Reinhart, T. (2000). The Theta System: Syntactic realization of verbal concepts. OTS Working Papers

In Linguistics.

Thompson, C. K. (2003). Unaccusative verb production in agrammatic aphasia: The argument structure complexity hypothesis. Journal of Neurolinguistics, 16, 151–167.

The comprehension of Catalan

OVS and OSV structures in Broca’s aphasia

Io Salmons and Anna Gavarró

Universitat Autònoma de Barcelona

Introduction

The aim of the present study is to test the comprehension of topicalisation in agrammatism in order to evaluate current hypotheses on the comprehension deficit in Broca’s aphasia. Such constructions present some interesting properties in Catalan that make them suitable for this purpose: they can present noncanonical word orders, which are predicted to be impaired by the Derived Order Problem Hypothesis (DOP-H; Bastiaanse & Van Zonneveld, 2005), and they are argued to be derived by movement, which is crucial for their analysis under the Trace-Deletion Hypothesis (TDH; Grodzinsky, 2000; 2006) and the Featural Underspecification Hypothesis (FUH; Grillo, 2008).

Antecedents

To our knowledge, there is only one study concerning the aphasic comprehension of Clitic Left Dislocations (CLLD) in Romance languages: Beretta et al. (2001), where the comprehension of two Spanish-speaking patients was tested. The mean performance of the two subjects was at chance; however, individual results reveal that one subject performed notably more successfully than the other (mean percentages of 75% and 35% of correct responses respectively). Hence, the sample was too small and the performance too heterogeneous to draw robust conclusions.

Methods

A binary sentence-picture matching task was conducted with nine Catalan-speaking agrammatic aphasics (4 females and 5 males, 1 left-handed, mean age of 59.6, all affected by CVA), and nine education- and age-matched control subjects. We tested 20 declaratives (1), 20 subject topicalisations (2), 20 CLLDs with preverbal subjects (3), 20 CLLDs with postverbal subjects (4) and 20 declaratives with a preverbal object clitic (5). All sentences were semantically reversible transitive sentences and were introduced by a sentence to justify the use of a topicalisation. All sentences were recorded digitally by a native speaker, so the characteristic intonational contour associated with topicalisation in Catalan was preserved.

(1) a. La pallasso va amagar als nens. Declarative (SVO)

D-f clown past-3s hide Acc+D-m.pl boy-pl ‘The clown hid the boys’

b. L’ àvia va pentinar a la nena. D-f.s grandmother past-3.s comb Acc D-f.s girl

‘The grandmother combed the girl’

(2) La pallasso, va amagar als nens. Subject topicalisation (SVO)

‘(As for) The clown, he hid the boys’

(3) Als nens, la pallasso els va amagar. CLLD with a preverbal subject (OSclV)

CL-m.pl

‘(As for) The boys, the clown hid them’

(4) Als nens, els va amagar la pallasso. CLLD with a postverbal subject (OclVS)

‘(As for) The boys, the clown hid them’

(5) La pallasso els va amagar. Declarative with an object clitic (SclV)

‘The clown hid them’

Results

Considering the results by item type (Table 1), the participants performed above chance in the control conditions, that is, declaratives and subject topicalisations. The mean percentage of correct responses reached 80.6% on the object clitics, which indicates that

this structure was not problematic for agrammatic individuals. The performance on CLLDs with preverbal subjects showed to be quite high as well, even though slightly worse than the performance on the control conditions. Finally, performance with CLLDs with postverbal subjects was at chance. For CLLD we looked at the impact that the resumptive clitic -in agreement with the subject in number and gender- could have had on the results, comparing the comprehension of those sentences where the subject and the object matched in gender and number (1b) to the comprehension of those sentences where there was a gender and number mismatch (1a). The aphasics’ performance did not depend on this factor: they were 63.3% correct with matched subject/object and 53.3% with the unmatched. These results are inconsistent with the FUH, according to which aphasics should benefit from mismatching of phi-features, as has been shown in other populations (Friedmann, Belletti and Rizzi, 2008).

Table 1. Mean percentages of correct responses by condition

SVO decl SVO top OSclV OclVS SclV

aphasic 90% 78.33% 67.78% 48.89% 80.6%

Discussion

The results reported here on declaratives and subject topicalisations are consistent with the hypotheses mentioned before, since such structures present the canonical word order in Catalan (SVO). The good performance shown in the comprehension of object clitics is in line as well with previous literature (Luzzatti et al. (2001) for Italian and Martínez-Ferreiro (2010) for Spanish, Galician and Catalan). Despite the noncanonical word order SclV of such structures, the theories that hypothesise the use of a linear strategy to assign thematic roles correctly predict the good performance reported in our study. Yet, it remains unclear whether the noncanonical position of the object clitic is predicted to be problematic under the DOP-H and the FUH. The performance on CLLDs with postverbals subjects (OclVS) dropped significantly, indicating that agrammatics fail in comprehending them, as predicted by all the hypotheses taken into consideration here. The same results have been reported in previous research on the comprehension of similar OVS structures, like focalisations (among others, Burchert et al., 2003;

Gavarró, 2005; Friedmann et al., 2010). However, agrammatics showed relatively unimpaired comprehension of CLLDs with preverbal subjects (OSclV), contrary to what has been reported in Beretta et al. (2001) for Spanish, and in studies where similar constructions were tested in other languages (Friedmann & Shapiro 2003 and

Friedmann et al. 2010 for Hebrew and Russian focalisations, respectively). Our findings challenge current theories on agrammatic comprehension, since these syntactic

constructions are argued to have undergone movement and present a noncanonical word order (OSclV) and, hence, they are expected to be problematic for agrammatic patients. All the hypotheses cited above fail in explaining the data reported in the

present study, which suggests that current theories on agrammatic comprehension need to be revised.

References

Bastiaanse, R. and R. Van Zonneveld (2005). Sentence production with verbs of

alternating transitivity in agrammatic Broca’s aphasia. Journal of Neurolinguistics,

18, 57–66.

Beretta, A., C. Schmitt, J. Halliwell, A. Munn, F. Cuetos and S. Kim (2001). The effects of scrambling on Spanish and Korean agrammatic interpretations: why linear models fail and structural models survive. Brain and Language, 79, 407–425. Burchert, F., R. De Bleser and K. Sonntag (2003). Does morphology make the difference?

Agrammatic sentence comprehension in German. Brain and Language, 87, 323– 342.

Friedmann, N., J. Reznick, D. Dolinski-Nuger and K. Soboleva (2010). Comprehension and production of movement-derived sentences by Russian-speaking agrammatic aphasics. Journal of Neurolinguistics, 23(1), 44–65.

Friedmann, N. and L. P. Shapiro (2003). Agrammatic comprehension of simple active sentences with moved constituents: Hebrew OSV and OVS structures. Journal of

Speech, Language, and Hearing Research, 46, 288–297.

Gavarró, A. (2005). The interpretation of ambiguous dislocations in Broca’s aphasia.

Brain and Language, 95, 117–118.

Grodzinsky, Y. (2000). The neurology of syntax: language use without Broca’s area.

Behavioral and Brain Sciencies, 23, 1–71.

Grodzinsky, Y. (2006). ‘A blueprint for a brain map of syntax’. In: Y. Grodzinsky and K. Amunts (Eds.), Broca’s Region. Oxford: Oxford University Press.

Luzzatti, C., A. Toraldo, M. T. Guasti, G. Ghirardi, L. Lorenzi and C. Guarnaschelli (2001). Comprehension of reversible active and passive sentences in agrammatism.

Aphasiology, 15, 419–441.

Martínez-Ferreiro, Silvia (2010). Towards a Characterization of Agrammatism in Ibero-

Romance. PhD dissertation. Universitat Autònoma de Barcelona.

The effects of slowed speech on comprehension of German

non-canonical sentences in aphasics with and without

hearing impairment

Angela Jochmann, Esther Ruigendijk Institute of Dutch Studies

Carl von Ossietzky University Oldenburg

Introduction

In aphasia, and especially in agrammatic aphasia, comprehension of non-canonical sentences as object-first structures is more impaired than that of canonical structures as subject-first sentence types. These difficulties increase with faster speech rates (Love et al., 2008). Aphasics often complain that people speak too fast (Silkes, 2012), and in aphasia therapy it is common to apply a slower speech rate. Studies have shown that agrammatic aphasics improve significantly in comprehending object-first structures when speech rate is slowed down (Dickey et al., 2007, Love et al., 2008). This leads to the idea that in

agrammatic aphasia the syntactic operations per se are not impaired, but that lexcial or syntactic processing is slowed down and that thus the time course of the acoustic presentation is too fast for agrammatic aphasics in order to compute the syntactic representations which results in comprehension problems (Burkhardt et al., 2003).

Many aphasics also experience deficits of intelligibility because of a hearing impairment, and are thus even further handicapped in understanding conversational speech as reduced intelligibility hinders comprehension as well. Slowing or time expansion of speech has been found to facilitate intelligibility and thus comprehension for spoken sentences in hearing impaired non-aphasic adults (Gordon-Salant et al., 2007).We therefore expect

time-expanded speech to improve intelligibility and comprehension in hearing impaired aphasics as well.

We hypothesize that slowed linguistic processing is responsible for the canonicity effect in agrammatic aphasia. Slower speech rates will result in better comprehension of non-canonical conditions for non-hearing impaired aphasics compared to performance at a normal speech rate. Hearing impaired aphasics will benefit from slowed speech in their performance in all structures.

Methods

In this study we examine the effects of uniformly time-expanded speech with four different German syntactic structures (two canonical and two non-canonical) and three different

speech rates- conversational, moderately expanded to 120% and highly expanded to 135% - on the comprehension performance of agrammatic and other aphasics with and without hearing impairment.

Design

We employ a sentence picture matching task with 15 items per syntactic structure: Two German main clauses (SVO, OVS) and two relative clause structures (SR, OR). These are presented in three different speech rates with two black and white pictures, one depicting the target and one serving as a distractor with reversed thematic roles. All subjects heard all 180 sentences.

Material

SVO Der kleine Junge umarmt den dicken Nikolaus.

TheNOM smallNOM boyNOM hugs theACC fatACC Santa Claus

‘The small boy hugs the fat Santa Claus.’

OVS Den dicken Nikolausi umarmt der kleine Junge _ti.

TheACC fatACC Santa Clausi hugs theNOM smallNOM boyNOM _ti.

‘It is the fat Santa Claus that the small boy hugs.’ SR Der Jungei, deri den Nikolaus umarmt, ist klein.

TheNOM boyNOMiwhoNOMi theACC Santa Claus hugs is small.

‘The boy, who hugs Santa Claus, is small.’

OR Der Jungei, deni der Nikolaus umarmt _ti, ist klein.

TheNOM boyNOMi whoACCi theNOM Santa ClausNOM hugs _ti is small.

‘The boy, who is being hugged by Santa Claus, is small.’

Participants

So far, 22 aphasics with mild to moderate severe impairment have been tested, 12 were diagnosed with agrammatic aphasia, 7 with hearing impairment. We also tested 35 age-matched controls with (N = 14) and without (N = 21) hearing impairment.

Results

Preliminary results as shown in Figure 1, show that all aphasics, and especially agrammatic aphasics, profit from slowed speech in their comprehension performance of non-canonical sentences. Hearing impaired aphasics benefit from slowed speech in their comprehension of

all structures. Non-aphasic hearing impaired controls also show improved comprehension when presented with slowed sentences.

Discussion

So far, our results show that in aphasia, slowed presentation facilitates comprehension of non-canonical sentences which might indicate that slowed language processing abilities are responsible for the deficit in comprehension of non-canonical sentences. Slowed speech also in general improves sentence comprehension both in hearing impaired aphasics and

controls, which is in line with Gordon-Salant et al. (2007).

These results indicate that providing more time compensates for auditory processing

problems caused by hearing impairment. They also confirm studies by Love et al. (2008) and Dickey et al. (2007) that more specific comprehension problems with non-canonical

sentences caused by syntactic processing problems can be compensated by slowed speech.

References

Burkhardt, P., Piñango, M. M., & Wong, K. (2003). The role of the anterior left

hemisphere in real-time sentence comprehension: Evidence from split intransitivity.

Brain and Language, 86(1), 9–22.

Dickey, M. W., Choy, J. J., & Thompson, C. K. (2007). Real-time comprehension of wh- movement in aphasia: Evidence from eyetracking while listening. Brain and

Language, 100(1), 1–22.

Gordon-Salant, S., Fitzgibbons, P. J., & Friedman, S. A. (2007). Recognition of time-

compressed and natural speech with selective temporal enhancements by young and elderly listeners. Journal of Speech, Language, and Hearing Research, 50, 1181–1193. Love, T., Swinney, D., Walenski, W., & Zurif, E. (2008). How left inferior frontal cortex

participates in syntactic processing: Evidence from aphasia. Brain and Language, 107, 203–219.

Silkes, J. P. (2012). Providing audiological services to individuals with aphasia:

Considerations, preliminary recommendations, and a call for research. American

Journal of Audiology, 21, 3–12.

The role of typological distance in differential impairments in

bilingual aphasia: evidence from Spanish-Basque

agrammatism

Amaia Munarriz1

1_{University of the Basque Country UPV/EHU}

Introduction

Research on bilingual aphasia provides evidence for one of the recurrent debates in

bilingualism, namely whether the two languages have a shared/separate representations in the bilingual brain (Abutalebi & Green, 2007; Paradis, 2004). On the one hand, several data from studies on bilinguals with aphasia are considered to present support in favor of the shared representation of the two languages: a) cases of parallel impairment and recovery (Abutalebi & Green, 2007); b) cross-linguistic transfer of therapy benefits from the treated to the untreated language (Faroqi-Shah, et al., 2010); c) cross-linguistic priming studies showing that hearing a sentence in one language can facilitate the production of a sentence with the same structure in another language (Verreyt, et al., 2013) and d) the finding that some bilinguals use the same processing strategy regarding cues for interpretation for their two languages (Wulfeck et al., 1986 ). On the other hand, data reporting non-parallel

impairments and recovery patterns in bilinguals with aphasia (Adrover-Roig et al., 2011; Kambanaros & Grohmann, 2012; Venkatesh et al., 2012) are considered to support the view on the neurofunctional separation of the two languages in the brain, and point towards distinct neural circuits within the same broad cortical areas (Paradis, 2004) casting doubts on the shared representation account.

Several factors are claimed to modulate different kinds of cross-linguistic influence (Faroqi-Shah et al., 2010; Goral et al., 2012) and consequently to affect postmorbid performance in bilingual aphasia (Ansaldo et al., 2008; Paradis, 2004). Among these factors, linguistic and structural distance appears as a promising one: the smaller the differences between the languages and the structures involved, the more likely it is for crosslinguistic influence to arise (cf. Verryt et al., 2013; Goral et al., 2010), and as a consequence the easier for parallell impairments to surface (De Diego Balaguer et al., 2004; Hernández et al., 2008; Tschirren et al., 2011), and the bigger the differences between the languages and the structures involved the less likely it is for CLE to surface and thus the more likely for non-parallel

impairment/recovery patterns to arise (cf. Diéguez-Vide et al., 2012; Goral et al., 2010; 2012; Kambanaros & Grohmann, 2012; Venkatesh et al., 2012).

In this study, we explore the morphosyntactic performance of a Spanish-Basque bilingual patient in her two languages in order to see whether the impairment affects the two languages similarly, and typological distance between these languages may influence the (dis)similarities found. According to finding from previous studies, a more similar

performance is expected across languages in those morphosyntactic structures which are at a surface level similar between the two languages (cf. Verreyt et al., 2013).

Methods

Participants

This study is based on a case study of an early Spanish-Basque bilingual with chronic Broca’s aphasia, which was highly proficient in her two languages premorbidly. The bilingual

participant received therapy in Spanish and after the lesion Spanish is almost exclusively the language used. A second unimpaired participant matched in gender, age, education and language background acted as a control.

Procedure

Spontaneous and experimental data from a variety of comprehension and production tasks, from more restrictive (picture description, sentence-to-picture matching) to more

naturalistic tasks (spontaneous speech, role playing) was collected 5 years post onset in order to assess performance of certain movement-derived structures in both languages. Three types of sentences were studied: a) canonical and topicalization root sentences, b) subject and object relatives and c) subject and object root questions.

Among these structures, canonical and topicalization structures as well as relatives differ in word order between Spanish and Basque, but questions have the same surface order in both languages.

Results

Results revealed important differences between tasks, modalities and languages. Although some target-deviant instances were attested in the production of questions in naturalistic tasks, the production of canonical and movement-derived structures appears to be preserved to a high extent as indicated by the near normal performance in picture

description experimental tasks in both languages (> 95% of accuracy). However, a selective and differential impairment surfaced in comprehension as revealed by sentence-picture-matching tasks: selective difficulties were attested in the comprehension of both questions and relatives (p < .001 both) (as well as in OSV and OSV topicalization structures depending on the task), and only in Basque (Figure 1).

Discussion

The preservation of movement derived structures in production might suggest a parallel impairment/recovery which could be due to some cross-language transfer of spontaneous or treatment-based recovery in production (cf. Goral et al., 2010). In contrast, the differential impairment affecting Basque more than Spanish not only in the structures that differ between the two languages (subject relatives) but also in those that have the same surface linear order (object questions), are difficult to explain if some strategy for comprehension were transferred cross-linguistically (cf. Venkatesh et al., 2012). We propose that several morphosyntactic differences between Spanish and Basque (head-final vs. head-initial, prenominal relatives vs. postnominal relatives, ergative vs. accusative overt marking) might have an impact on the different reliability and availability of morphosyntactic cues for comprehension in both languages. Thus, we claim that morphosyntactic distance between

Spanish and Basque might have hindered the possible cross-linguistic transfer of

comprehension strategies from the best preserved Spanish to the lesser preserved Basque. This interpretation is in line with other cases of non-parallel recovery in contact situations of typologically distant language pairs (cf. Diéguez-Vide et al., 2012; Kambanaros & Grohmann, 2012; Venkatesh, et al., 2012).

Finally, the differential impairment attested in this bilingual with aphasia challenges neuropsycholinguistic models which argue in favor of shared syntactic representations (Verreyt et al., 2013, see also Ullman, 2001 for early bilinguals) and appear to be more compatible the view on the neurofunctional isolability of the morphosyntactic

representations of the two languages (Paradis, 2004). Thus, our data along the lines of other studies indicate that typological distance between languages in general, and language-specific features of the structures in particular surface as key factors for

impairment/recovery patterns in bilingual aphasia.

References

Author, X. Y. (2012). Title of paper. Title of Journal, 1, 1-2.

Abutalebi, J., & Green, D. W. (2007). Bilingual language production: The neurocognition of language representation and control. Journal of Neurolinguistics, 20, 242-275.

Adrover-Roig, D., Galparsoro-Izagirre, N., Marcotte, K., Ferré, P., Wilson, M. A., & Ansaldo, A. I. (2011). Impaired L1 and executive control after left basal ganglia damage in a bilingual Basque-Spanish person with aphasia. Clinical Linguistics and Phonetics, 25, 480-498.

De Diego Balaguer, R., Costa, A., Sebastián-Gallés, N., Juncadella, M., & Caramazza, A. (2004). Regular and irregular morphology and its relationship with agrammatism: Evidence from two Spanish-Catalan bilinguals. Brain and Language, 91, 212-222. Diéguez-Vide, F., Gich-Fullà, J., Puig-Alcántara, J., Sánchez-Benavides, G., & Peña-Casanova, J.

(2012). Chinese-Spanish-Catalan trilingual aphasia: a case study. Journal of

Neurolinguistics, 25, 630-641.

Faroqi-Shah, Y., Frymark, T., Mullen, R., & Wang, B. (2010). Effect of treatment for bilingual individuals with aphasia: a systematic review of the evidence. Journal of

Neurolinguistics, 23, 319-341.

Goral, M., Levy, E. S., & Kastl, R. (2010). Cross-language treatment generalisation: A case of trilingual aphasia. Aphasiology, 24, 170-187.

Goral, M., Rosas, J., Conner, P. S., Maul, K. K., & Obler, L. K. (2012). Effects of language proficiency and language of the environment on aphasia therapy in a multilingual.

Journal of Neurolinguistics, 25, 538-551.

Hernández, M., Caño, A., Costa, A., Sebastián-Gallés, N., Juncadella, M., & Gascón-Bayarri, J. (2008). Grammatical category-specific deficits in bilingual aphasia. Brain and

Language, 107, 68-80.

Kambanaros, M., & Grohmann, K. K. (2012). BATting multilingual primary progressive aphasia for Greek, English, and Czech. Journal of Neurolinguistics, 25, 520-537. Paradis, M. (2004). A neurolinguistic theory of bilingualism. Amsterdam/Philadelphia: John

Benjamins.

Tschirren, M., Laganaro, M., Michel, P., Martory, M.-D., Di Pietro, M., Abutalebi, J., et al. (2011). Language and syntactic impairment following stroke in late bilingual aphasics.

Brain and Language, 119, 238-242.

Ullman, M. T. (2001). The neural basis of lexicon and grammar in first and second language: the declarative/procedural model. Bilingualism: Language and Cognition, 4, 105-122. Venkatesh, M., Edwards, S., & Saddy, J. D. (2012). Production and comprehension of English

and Hindi in multilingual transcortical aphasia. Journal of Neurolinguistics, 25, 615-629.

Verreyt, N., Bogaerts, L., Cop, U., Bernolet, S., De Letter, M., Hemelsoet, D., et al. (2013). Syntactic priming in bilingual patients with parallel and differential aphasia.

Aphasiology, 27, 867-887.

Wulfeck, B. B., Juarez, L., Bates, E. A., & Kilborn, K. (1986). Sentence interpretation strategies in healthy and aphasic bilingual adults. In J. Vaid (Ed.), Language processing in

bilinguals (pp. 199-219). Hillsdale, NJ: Lawrence Erlbaum Associates.

Figure 1. AF’s accuracy percentages in the comprehension of root sentences, relatives and questions in sentence-picture matching tasks.

97.5% 98% 100.0% _95.2% 100.0% 100.0% 32% 81% 100% 50% 95.0% 100% 0% 20% 40% 60% 80% 100% Canonical O(V)S(V) SR OR SQ OQ

Root sentences Relatives Questions

A ccu ra cy ( % ) Spanish Basque 22

Do Pronouns Make a Difference?

On-line Processing of Relative Clauses in the Visual-world

Paradigm

Anne Adelt, Nicole Stadie, Frank Burchert1

1_{Department of Linguistics, University of Potsdam, Germany}

Introduction

Considering the processing of relative clauses, cross-linguistic studies in aphasia have often revealed patterns of above chance performance for subject relative clauses (SRCs) and chance performance for object relative clauses (ORCs) in off-line tasks such as sentence-picture verification (e.g., Burchert, De Bleser, & Sonntag, 2003). A similar subject-object asymmetry was observed on-line, as evidenced by longer self-paced listening times for ORCs compared to SRCs (Caplan, Waters, Dede, Michaud, & Reddy, 2007).

In an attempt to explain sentence comprehension deficits in aphasia, Garraffa and Grillo (2008) and Grillo (2008) proposed an extension of the linguistic framework of Relativized Minimality (RM; Rizzi, 2004). They argued that movement-derived non-canonical sentences (such as ORCs) cause a processing disadvantage compared to SRCs, because only the former contains more than one potential antecedent for the trace in the extraction site of the object. Interestingly, Friedmann et al. (2009) observed that in Hebrew typically-developing children this processing disadvantage for ORCs disappears when one of the antecedents is a pronoun.

The present ongoing study focuses on German, a language in which SRCs and ORCs are strict minimal pairs in terms of word order. We adopt the extension of RM and investigate

whether the presence of pronouns facilitates the processing of ORCs in healthy adults and individuals with aphasia (IWA). To our knowledge, such an investigation has not been published yet.

Methods

Material

Test sentences are 64 RCs in total, 32 SRCs and 32 ORCs. All RCs contain an RC head which is followed by a second constituent that is either a DP or a personal pronoun. RC heads contain eight singular feminine or eight singular neuter nouns, whereas all the second constituents are masculine and marked for plural. Due to case syncretism, RC head and second

constituent provide no information about word order. Disambiguation occurs sentence finally at the finite verb through number marking. A sentence final verb in singular is in agreement with the RC head and refers to an SRC, while a final verb in plural agrees with the second constituent and refers to an ORC. Accordingly, target sentences are distributed over four conditions with 16 sentences each: SRC with two full DPs (1), SRC with a pronoun as second constituent (2), ORC with two full DPs (3), ORC with a pronoun as second constituent (4). Moreover, 32 questions with a prepositional phrase referring to a symbol that identifies one of the animals (see balloon on pig A in Figure 1) were used as fillers (5).

Relative clause RC head Subject extraction site 2nd constituent Object extraction site Final verb SRC (1) full DP,

n=16 Wo ist das Schwein+NP

dasi ti +NP die Wölfe+NP kitzelt

(Where is the pig that is tickling the wolves?) (2) pronoun,

n=16 Wo ist das Schwein+NP

dasi ti +NP sie-NP kitzelt

(Where is the pig that is tickling them?) ORC (3) full DP,

n=16 Wo ist das Schwein+NP

dasi die Wölfe+NP ti +NP kitzeln

(Where is the pig that the wolves are tickling?) (4) pronoun,

n=16 Wo ist das Schwein+NP

dasi sie-NP ti +NP kitzeln

(Where is the pig that they are tickling?) (5) Filler,

n=32 Wo ist das Schwein mit dem Ballon (Where is the pig with the balloon?)

Procedure

All sentences are presented auditorily and animated colored illustrations of animals are shown simultaneously on a computer screen. In all sentences, animals are arranged as follows: One animal (X) is on the left side of the screen, two other animals (YY) are in the middle, and one animal is placed on the right side of the screen (see an example in Figure 1). Animals and the performance of the action (e.g., tickling) are introduced in a preview

followed by the auditory presentation of the target sentences. The participants’ task is to identify the animal X to which the sentence refers. During this task, eye movements are collected as an on-line measure of sentence processing. Off-line comprehension is measured in terms of accuracy of target identification.

Participants

Ten IWAs with sentence comprehension deficits that are not due to pre-lexical and lexical impairments take part in the experiment. Additionally, 30 healthy controls are tested, matched in age and years of education.

Hypothesized Outcomes

Following the RM approach, our predictions are as follows:

(i) SRC vs. ORC: The performance for all SRCs irrespective of the presence of a pronoun should be different compared to ORCs in terms of higher accuracy scores in IWAs’ off-line comprehension and faster or longer fixations to the target picture in both

participant groups.

(ii) ORC in controls: ORCs with a personal pronoun as a second constituent (cf. 4) are assumed to give rise to a processing advantage (at least on-line; see Gordon, Hendrick, & Johnson, 2001 for similar results in self-paced reading experiments) compared to ORCs with a DP as a second constituent (cf. 3). There should be no such difference in their off-line performance.

(iii) ORC in IWA: A processing advantage for ORC with pronouns as a second constituent (cf. 4) is assumed to emerge both on- and off-line in contrast to ORC with DPs as a second constituent (cf. 3).

Results and discussion

Data collection and analysis is still ongoing. Results will be ready by September and

presented and discussed with respect to the above mentioned hypotheses at the Science of Aphasia conference.

References

Burchert, F., De Bleser, R., & Sonntag, K. (2003). Does morphology make the difference? Agrammatic sentence comprehension in German. Brain and Language, 87, 323–342.

Caplan, D., Waters, G., Dede, G., Michaud, J., & Reddy, A. (2007). A study of syntactic processing in aphasia I: Behavioral (psycholinguistic) aspects. Brain and Language, 101, 103–150.

Garraffa, M., & Grillo, N. (2008). Canonicity effects as grammatical phenomena. Journal of

Neurolinguistics, 21, 177–197.

Friedmann, N., Belletti, A., & Rizzi, L. (2009). Relativized relatives: Types of intervention in the acquisition of A-bar dependencies. Lingua, 119, 67–88.

Gordon, P. C., Hendrick, R., & Johnson, M. (2001). Memory interference during language processing.

Journal of Experimental Psychology: Learning, Memory, and Cognition, 27, 1411–1423.

Grillo, N. (2008). Generalized Minimality: Syntactic underspecification in Broca’s aphasia. PhD dissertation distributed by LOT, University of Utrecht, the Netherlands.

Rizzi, L. (2004). Locality and Left Periphery. In A. Belletti (Ed.), Structures and Beyond: The

Cartography of Syntactic Structures, Volume 3 (pp. 223–251). New York: Oxford University

Press.

Figure 1. Sample picture for sentences 1-5. Target in SRCs and filler: pig (A), target in ORCs: pig (B)